Energy-saving aquatic product ultralow-temperature cold chain transportation device and preservation method

Abstract

The invention relates to the technical field of cold-chain transportation, and discloses an energy-saving aquatic product ultralow-temperature cold-chain transportation device and a fresh-keeping method.The energy-saving aquatic product ultralow-temperature cold-chain transportation fresh-keeping method comprises the steps that aviation cargo hold direct-current power source data is obtained, and an ultralow-temperature environment data set is generated through direct-current boost conversion and three-stage cascade refrigeration; constructing an energy storage layer and controlling intermittent operation; magneto-rheological vibration reduction is carried out, and a cooperative control strategy is combined; defrosting collaborative multispectral sensing and quality prediction are carried out; a hierarchical control strategy of a three-layer architecture is adopted to obtain a multi-parameter cooperative control scheme; performing thermoelectric conversion; performing multi-layer composite heat preservation to obtain a heat preservation structure optimization scheme; under the condition that air transportation is limited, the ultralow temperature of minus 65 DEG C can be stably kept, temperature fluctuation is controlled within an extremely small range, aquatic product tissues are effectively protected against vibration damage, the preservation effect is monitored in real time, and a technical solution is provided for high-end aquatic product air cold chain transportation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cold chain transportation, more particularly, it relates to an energy-saving type super-low-temperature cold chain transportation device for aquatic products and a preservation method. BACKGROUND

[0002] With the rapid development of global aquatic product trade, the cross-continental transportation demand of high-end aquatic products such as bluefin tuna and northern shrimp is increasing. Such aquatic products are rich in myoglobin and unsaturated fatty acids, and are extremely sensitive to temperature, and must be transported in a super-low-temperature environment to maintain the best quality. Bluefin tuna, as the top sashimi material, requires to be stored in a super-low-temperature environment below minus sixty degrees Celsius to inhibit myoglobin oxidation and lipid peroxidation reactions, and maintain the fresh red color and delicate taste of the meat. However, the traditional super-low-temperature cold chain transportation method faces many challenges in the aviation transportation scenario.

[0003] Although the traditional liquid nitrogen cold chain transportation method can provide a super-low-temperature environment, it has problems such as difficulty in accurately controlling the temperature due to continuous evaporation of liquid nitrogen, large temperature fluctuation range, difficulty in supplementing liquid nitrogen, and high transportation cost. Liquid nitrogen continuously evaporates during transportation, and may be completely depleted after long-distance transportation, resulting in temperature loss of control. At the same time, liquid nitrogen is a dangerous product and is strictly limited in aviation transportation, increasing the complexity and cost of transportation. Although the ordinary mechanical refrigeration method can be recycled, single-stage or double-stage compression refrigeration systems cannot reach minus 65 degrees Celsius in the limited power supply conditions of the aviation cargo cabin, and the insufficient refrigeration capacity leads to deterioration of the quality of aquatic products. In addition, the special environmental conditions such as air pressure changes and severe vibrations experienced during aviation transportation put high requirements on the stability, vibration resistance and energy saving of the super-low-temperature cold chain transportation device.

[0004] The aviation cargo cabin environment is special, the power supply is usually 28-volt direct current power and the power is strictly limited, and during transportation, it experiences multiple stages such as ground standby, take-off climb, cruise, descent and landing, and the environmental air pressure decreases from ground standard atmospheric pressure to low air pressure environment at cruising altitude, and the peak value of vibration acceleration during take-off and landing can reach several times of gravity acceleration. These factors bring severe challenges to the refrigeration system performance, temperature stability, energy consumption control, vibration protection, etc. of the super-low-temperature cold chain transportation device. The existing super-low-temperature cold chain transportation device cannot meet the requirements of super-low-temperature preparation, temperature accurate control, low energy consumption operation, vibration protection, quality real-time monitoring, etc. at the same time, which limits the development of high-end aquatic product aviation cold chain transportation.

[0005] Therefore, it is urgent to develop an energy-saving type super-low-temperature cold chain transportation device for aquatic products and a preservation method suitable for aviation transportation scenarios to meet the actual needs of cross-continental transportation of high-end aquatic products. SUMMARY

[0006] The application provides an energy-saving aquatic product ultra-low-temperature cold chain transportation device and preservation method, and solves the technical problems of being difficult to simultaneously meet ultra-low-temperature preparation, temperature accurate control, low-energy-consumption operation, vibration protection and quality real-time monitoring in the related art.

[0007] The application provides an energy-saving aquatic product ultra-low-temperature cold chain transportation preservation method, which comprises the following steps: Obtaining aviation cargo cabin direct-current power supply data, generating an ultra-low-temperature environment data set through direct-current voltage conversion and three-stage cascade refrigeration, Based on the ultra-low-temperature environment data set, an energy storage layer is constructed and intermittent operation is controlled to generate an optimized ultra-low-temperature environment data set; Based on the optimized ultra-low-temperature environment data set, a magnetic rheological vibration reduction is performed and a collaborative control strategy is combined to obtain a vibration reduction protection control scheme; Based on the aquatic products in the vibration reduction protection control scheme, defrosting, collaborative multi-spectrum sensing and quality prediction are performed to generate a quality monitoring data set; Based on the quality monitoring data set, the optimized ultra-low-temperature environment data set and the ultra-low-temperature environment data set, a three-layer hierarchical control strategy is adopted to obtain a multi-parameter collaborative control scheme; Based on the multi-parameter collaborative control scheme, a thermoelectric conversion is performed to obtain an energy recycling scheme; Based on the energy recycling scheme, a multi-layer composite thermal insulation is performed to obtain a thermal insulation structure optimization scheme.

[0008] In a preferred embodiment, the generating of the ultra-low-temperature environment data set comprises: Obtaining aviation cargo cabin direct-current power supply data, and adopting a direct-current voltage conversion module to convert the voltage of the aviation cargo cabin direct-current power supply to obtain a compressor power supply; Designing a three-stage cascade refrigeration system, a low-temperature stage refrigeration circuit adopts a first refrigerant, and the evaporation temperature of a low-temperature stage evaporator is set as a first temperature threshold; a medium-temperature stage refrigeration circuit adopts a second refrigerant, and a medium-temperature stage evaporator simultaneously serves as a low-temperature stage condenser; and a high-temperature stage refrigeration circuit adopts a third refrigerant, and a high-temperature stage evaporator simultaneously serves as a medium-temperature stage condenser; The in-box evaporator is connected to the low-temperature stage refrigeration circuit, a low-temperature axial flow fan is configured to form air circulation, a temperature sensor is arranged to monitor the temperature, and the three-stage compressors are started in a preset order, and the temperature in the box is reduced to a second temperature threshold.

[0009] In a preferred embodiment, the generating of the optimized ultra-low-temperature environment data set comprises: Preparing a composite phase change material, selecting a low-temperature eutectic mixture as a phase change base material, adding a nano nucleating agent to reduce the supercooling degree of the phase change process, adding a thermal conductivity enhancer to improve the thermal conductivity, and adjusting the component ratio to control the phase change temperature point at a third temperature threshold; The composite phase change material is packaged and installed on the inner wall of the box, a honeycomb plate is used as a packaging container, the composite phase change material is injected into the honeycomb cavity, heat conduction fins are arranged in the honeycomb cavity to enhance heat transfer, and a phase change energy storage layer is formed; The phase change energy storage layer is used to control the intermittent operation control strategy of the refrigeration system; the three-stage compressor is operated in the refrigeration stage to reduce the temperature of the box and store the cold energy of the phase change material; the compressor is turned off in the heat preservation stage to release the cold energy of the phase change material to maintain the temperature stable, and the temperature fluctuation is controlled within a preset range.

[0010] In a preferred embodiment, the obtained vibration damping protection control scheme comprises: A magneto-rheological damper is installed at the bottom of the box, including an outer cylinder, a piston rod, a piston head, a magneto-rheological fluid cavity and an excitation coil, the apparent viscosity of the magneto-rheological fluid is changed by adjusting the excitation current, and the damping coefficient adjustment range is within a preset range; A three-axis acceleration sensor is used to collect vibration signals and calculate the optimal damping force, a semi-active control algorithm is used to numerically integrate the acceleration signals, the expected damping force is calculated according to the skyhook damping control law, and the required excitation current is calculated through the mechanical model of the damper; The root mean square value of the vibration acceleration is calculated in real time, when the root mean square value of the vibration exceeds the first vibration threshold, a cooling instruction is sent to the temperature control system to reduce the temperature of the box to the fourth temperature threshold, and the peak value of the vibration acceleration is controlled below the second vibration threshold.

[0011] In a preferred embodiment, the generated quality monitoring data set comprises: A multi-spectral sensor is installed in the box and a defrosting device is configured, the multi-spectral sensor includes a near-infrared spectrometer and a visible light spectrometer, the probe surface temperature is controlled in a pulse heating mode, and spectrum collection is performed in a defrosting window period; The surface spectrum data of the aquatic products are collected, the light source and the spectrometer are triggered to collect the reflectance spectrum, the reflectivity of the characteristic wavelength and the absorbance of the characteristic wavelength are extracted, and the spectrum characteristic parameters are obtained; A partial least squares regression prediction model is used, the spectrum characteristic parameters are used as independent variables, the myoglobin oxidation degree and the lipid peroxide value are used as dependent variables, a linear regression relationship is established, and real-time quality index data are obtained.

[0012] In a preferred embodiment, the obtained multi-parameter collaborative control scheme comprises: A fuzzy control algorithm is used, the input is the temperature deviation and the deviation change rate in the box, and the output is the compressor frequency increment, and the control period is within a first time threshold; A model predictive control algorithm is used, the input includes the temperature distribution in the box, the environmental air pressure and the quality index, and the output is the compressor frequency set value and the expansion valve opening degree set value, a system prediction model and an optimization objective function are established; A deep reinforcement learning algorithm running on a cloud server is used to establish an actuator network and an evaluator network, and the control parameters are updated through offline pre-training and online optimization.

[0013] In a preferred embodiment, the energy recovery utilization scheme includes: A heat exchange copper plate is installed, and thermoelectric conversion modules are connected in series, with the hot end of the thermoelectric conversion modules attached to the copper plate and the cold end attached to a heat dissipation plate. When there is a temperature difference between the hot end and the cold end, the thermoelectric conversion module outputs a direct current voltage and power, and the output voltage is reduced and stabilized to a preset voltage through a direct current converter. The heat dissipation strategy is dynamically adjusted according to the output power of the thermoelectric conversion module, and when the output power is below a first power threshold, forced cooling is started, and when the output power recovers to above a second power threshold, forced cooling is reduced or turned off to maintain stable temperature difference.

[0014] In a preferred embodiment, the insulation structure optimization scheme includes: A multi-layer composite insulation system is designed, which adopts a four-layer composite structure of shell, vacuum insulation board, aerogel felt and inner wall, the vacuum insulation board is composed of core material, gas barrier film and vacuum packaging, the core material adopts fumed silica, and the aerogel felt is laid on the inner side of the vacuum insulation board. The heat transfer coefficient is calculated and the heat bridge treatment is optimized, the overall heat transfer coefficient is calculated using a heat transfer series resistance model, and the support column is optimized and designed as a I-shaped section with an aerogel filling cavity. A sealing structure is designed, an opening door is provided at the top of the box, a double-layer sealing structure is adopted between the door and the box, the inner sealing ring adopts silicone rubber material, the outer sealing adopts polytetrafluoroethylene sealing tape, and the door body locking adopts a multi-point lock, and the overall heat transfer coefficient of the box is controlled below a preset threshold.

[0015] In a preferred embodiment, the design of the three-stage cascade refrigeration system further includes: The first refrigerant of the low-temperature stage refrigeration circuit is azeotropic mixture, with normal boiling point below the fifth temperature threshold, capable of realizing evaporation temperature below the sixth temperature threshold; The low-temperature stage compressor, the medium-temperature stage compressor and the high-temperature stage compressor all adopt direct current variable frequency compressors, and the speed is adjusted through a frequency controller, and each stage circuit is configured with an electronic expansion valve to adjust the refrigerant flow, and the valve opening adjustment range is within a preset range; A barometric pressure sensor is installed outside the box, a barometric pressure refrigeration power compensation model is established, and when the detected barometric pressure is reduced to below a first barometric pressure threshold, the condenser fan speed and the compressor frequency are increased to compensate for the influence of barometric pressure.

[0016] The application provides an energy-saving aquatic product ultra-low-temperature cold chain transportation device for performing the energy-saving aquatic product ultra-low-temperature cold chain transportation and preservation method. A three-stage cascade refrigeration module is used to obtain air cargo cabin direct current power supply data, generate an ultra-low-temperature environment data set through direct current boost conversion and three-stage cascade refrigeration. A phase change energy storage module is used to build an energy storage layer and control intermittent operation based on the ultra-low-temperature environment data set, and generate an optimized ultra-low-temperature environment data set. A magneto-rheological vibration reduction module is used to perform magneto-rheological vibration reduction and combine a cooperative control strategy based on the optimized ultra-low-temperature environment data set, and obtain a vibration reduction protection control scheme. A multi-spectral quality monitoring module is used to perform defrosting cooperative multi-spectral sensing and quality prediction based on aquatic products in the vibration reduction protection control scheme, and generate a quality monitoring data set. A multi-parameter cooperative control module is used to obtain a multi-parameter cooperative control scheme by adopting a three-layer architecture hierarchical control strategy based on the quality monitoring data set, the optimized ultra-low-temperature environment data set and the ultra-low-temperature environment data set. A thermoelectric conversion energy recovery module is used to perform thermoelectric conversion based on the multi-parameter cooperative control scheme, and obtain an energy recovery utilization scheme. A vacuum thermal insulation module is used to perform multi-layer composite thermal insulation based on the energy recovery utilization scheme, and obtain a thermal insulation structure optimization scheme.

[0017] The application has the following beneficial effects: By adopting the multi-technology integrated scheme of the three-stage cascade refrigeration system adapted to the aviation power supply, the composite phase change energy storage system reinforced by the nano nucleating agent, the magneto-rheological intelligent vibration reduction system, the defrosting cooperative multi-spectral quality monitoring system, the hierarchical architecture multi-parameter cooperative temperature control strategy, the thermoelectric conversion energy recovery system and the vacuum thermal insulation system, the problem of ultra-low-temperature preservation of high-end aquatic products under the limited conditions of aviation transportation is effectively solved. The zero-65-degree Celsius ultra-low-temperature environment can be stably maintained for a long time under the limited power supply conditions of the aviation cargo cabin, the temperature fluctuation is controlled within a very small range, and the high-end aquatic products such as bluefin tuna can maintain the quality of sashimi during long-distance transportation. The phase change energy storage system realizes intermittent operation of the refrigeration system, reduces energy consumption, and the energy consumption cost is much lower than that of the traditional liquid nitrogen transportation method. The magneto-rheological vibration reduction system combines the vibration-temperature cooperative control strategy, effectively protects the aquatic product tissue from damage caused by severe vibration during aviation transportation. The multi-spectral quality monitoring system can evaluate the preservation effect in real time, and provide reliable quality guarantee for the transportation process.

[0018] It has good environmental adaptability and long-term operation stability, and can work normally under extreme conditions such as high temperature, low temperature, low pressure and strong vibration. The phase change energy storage system keeps stable performance after multiple cycles, the magneto-rheological damper responds quickly and the damping force adjustment range does not decay, the prediction accuracy of the spectrum monitoring system remains at a high level, and the thermal conductivity of the vacuum insulation panel does not increase obviously. It provides an ideal technical solution for transpacific air cold chain transportation of high-end aquatic products, and has obvious technical progress and practical value. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is the main flow chart of an energy-saving type aquatic product ultra-low temperature cold chain transportation and preservation method of the present application; Figure 2 is a detailed flow chart of an energy-saving type aquatic product ultra-low temperature cold chain transportation and preservation method of the present application Figure 3 is a module diagram of an energy-saving type aquatic product ultra-low temperature cold chain transportation and preservation system of the present application. DETAILED DESCRIPTION

[0020] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that the discussion of these implementations is merely meant to provide a better understanding of the subject matter described herein and can be changed in function and arrangement without departing from the scope of the content of the present specification. Various processes or components can be omitted, replaced or added according to needs of various examples. In addition, features described in some examples can also be combined in other examples.

[0021] In at least one embodiment of the present application, an energy-saving type aquatic product ultra-low temperature cold chain transportation and preservation method is disclosed, as shown in Figures 1 to 2 , comprising: Step 1, obtaining the data of the direct current power supply of the air cargo cabin, generating the data set of the ultra-low temperature environment through direct current boost conversion and three-stage cascade refrigeration; Specifically comprising the following steps: Step 1.1, obtaining the data of the 28-volt direct current power supply of the air cargo cabin and performing voltage conversion; Based on the 28-volt direct current power supply data output by the air cargo cabin power distribution system, an isolation type DC-DC boost conversion module is used to boost the input voltage, and a 48-volt direct current power supply is obtained for subsequent use of the compressor. In one embodiment, the DC-DC boost conversion module uses a full-bridge LLC resonant topology structure, which works under the condition of zero-voltage opening of the switching tube through the resonant cavity, reduces the switching loss, the input voltage range is 24 volts to 32 volts, the output voltage is stabilized at 48 volts, the peak-to-peak value of the ripple voltage is less than 100 millivolts, and the conversion efficiency is above 95%.

[0022] To ensure the quality of power supply, the module is internally integrated with an input filter capacitor bank, which is configured in parallel with an aluminum electrolytic capacitor and a ceramic capacitor, with a total capacitance of 4000 microfarads, to suppress input current ripple and prevent interference with the aviation power supply system. The boost module output is connected to an output filter inductor and capacitor, with an inductance of 200 microhenries and a filter capacitance of 2000 microfarads, to ensure stable output voltage.

[0023] In terms of safety protection, the boost module is configured with overcurrent protection, which automatically limits the current when the output current exceeds 65 amperes, and overtemperature protection, which reduces the output or shuts down when the module temperature exceeds 85 degrees Celsius.

[0024] Step 1.2, design a three-stage cascade refrigeration system based on a 48-volt DC power supply; Based on a 48-volt DC power supply, a three-stage cascade refrigeration system is used to achieve ultra-low temperature refrigeration capability of minus 65 degrees Celsius. Specifically, the low-temperature stage refrigeration circuit uses R508B refrigerant, which is a azeotropic mixture of R23 and R116, with a normal boiling point of minus 86.2 degrees Celsius, capable of achieving an evaporation temperature of minus 90 degrees Celsius. The low-temperature stage compressor is a direct-current variable-frequency rotary compressor with a rated power of 1000 watts and a frequency adjustment range of 30 to 90 hertz. The frequency range is selected based on the following criteria: below 30 hertz, the compressor lubrication is poor and the refrigeration capacity is too small; above 90 hertz, the mechanical stress is too large and the energy efficiency ratio decreases. The low-temperature stage evaporator uses a copper tube and aluminum fin structure with a tube diameter of 6 mm, a fin pitch of 3 mm, and a heat exchange area of 0.8 square meters. The evaporation temperature is set to minus 75 degrees Celsius, and the evaporation pressure is 0.2 MPa gauge.

[0025] The medium-temperature stage refrigeration circuit uses R404A refrigerant, which is a mixture of R125, R143a, and R134a, with a normal boiling point of minus 46.5 degrees Celsius. The medium-temperature stage compressor is a direct-current variable-frequency scroll compressor with a rated power of 1200 watts and a frequency adjustment range of 30 to 90 hertz. The medium-temperature stage evaporator also serves as the low-temperature stage condenser, using a plate heat exchanger structure with a heat exchange area of 0.5 square meters. The low-temperature stage refrigerant condenses in the channel between the plates at a temperature of minus 50 degrees Celsius and a pressure of 0.8 MPa gauge. The medium-temperature stage refrigerant evaporates in the other channel at a temperature of minus 55 degrees Celsius and a pressure of 0.15 MPa gauge.

[0026] The high-temperature stage refrigeration circuit uses R134a refrigerant, with a normal boiling point of minus 26.1 degrees Celsius. The high-temperature stage compressor is a direct-current variable-frequency rotor compressor, with a rated power of 800 watts and a frequency adjustment range of 30 to 90 Hz. The high-temperature stage evaporator also serves as the medium-temperature stage condenser, adopting a double-pipe heat exchanger structure with an inner tube diameter of 12 mm, an outer tube diameter of 20 mm, and a heat exchange length of 5 m. The medium-temperature stage refrigerant condenses in the inner tube at a condensing temperature of minus 20 degrees Celsius and a condensing pressure of 1.2 MPa gauge. The high-temperature stage refrigerant evaporates in the annular space between the inner and outer tubes at an evaporating temperature of minus 25 degrees Celsius and an evaporating pressure of 0.1 MPa gauge.

[0027] The high-temperature stage condenser is a forced air-cooled finned tube heat exchanger with a heat exchange area of 1.2 m2, a condensing temperature of 40 degrees Celsius, and a condensing pressure of 1.02 MPa gauge. It is equipped with an axial flow fan with a wind volume of 400 m3 / h and a power of 50 W. To achieve precise control, electronic expansion valves are installed in each stage of the three-stage refrigeration system to regulate the refrigerant flow. The valve opening adjustment range is 10% to 100%, with a response time of less than 1 second. The control system dynamically adjusts the valve opening based on the evaporating pressure and superheat of each stage.

[0028] Step 1.3: Establishing an ultra-low temperature environment in the box based on the three-stage cascade refrigeration system; Based on the refrigeration capacity output by the three-stage cascade refrigeration system, an air circulation heat exchange method is used to achieve a uniform and stable ultra-low temperature environment of minus 65 degrees Celsius in the box. In one embodiment, an in-box evaporator is installed at the central top position of the box. The evaporator is connected to the low-temperature stage refrigeration circuit and adopts a copper tube with aluminum fins structure, with a tube diameter of 8 mm and a fin spacing of 5 mm. The heat exchange area is 1 m2, and the refrigerant R508B evaporates in the tube to absorb heat. The evaporating temperature is minus 75 degrees Celsius. The air in the box is cooled by the evaporator and naturally sinks due to increased density, forming a natural convection circulation. To enhance the circulation, a low-temperature axial flow fan is installed below the evaporator. The fan is designed to withstand low temperatures, with a brushless DC motor, a power of 30 W, a wind volume of 100 m3 / h, and a blade material of carbon fiber reinforced polyether ether ketone, with a temperature resistance range of minus 100 to 200 degrees Celsius.

[0029] To achieve precise temperature monitoring, eight temperature sensors are arranged at the four corners of the box, both above and below. The sensors use platinum resistance PT1000 type, with a temperature measurement range of minus 100 to 100 degrees Celsius and an accuracy level of A class. The measurement error is less than 0.2 degrees Celsius at minus 65 degrees Celsius. The sensor signals are connected to the temperature acquisition module through a four-wire connection method to eliminate the influence of lead resistance.

[0030] After the refrigeration system is started, the three-stage compressor is started in the order of low-temperature stage, medium-temperature stage and high-temperature stage, and the starting interval is 3 minutes to avoid instantaneous power over-limit. In the initial stage, the three-stage compressor is operated at a medium frequency of 50 Hz, the temperature in the box body starts to drop from the initial ambient temperature of 25 DEG C, the temperature drop rate is about 1 DEG C per minute at 0 DEG C or above, the temperature drop rate gradually decreases below 0 DEG C, and the temperature drop rate is about 0.3 DEG C per minute below -30 DEG C. After about 150 minutes, the temperature in the box body reaches -65 DEG C and stabilizes, at this time the total power of the three-stage compressor is 2800 W, and the system energy efficiency ratio COP is 0.85, that is, 1000 W of electric power can produce 850 W of refrigeration capacity.

[0031] In another embodiment, since the air transport environment air pressure change will affect the refrigeration system performance, an air pressure compensation control strategy can be used, the purpose is to maintain the refrigeration capacity stable under different air pressure conditions, preferably, an air pressure sensor is installed outside the box body, the measurement range is 70-110 kPa, the accuracy is 0.5 kPa, the air pressure signal is connected to the control system, an air pressure-refrigeration power compensation model is established, when the air pressure is detected to decrease from the ground 101 kPa to the cruising altitude 75 kPa, since the air density decreases, the high-temperature stage condenser heat exchange efficiency decreases by about 15%, through the model calculation, it is obtained that the condensing fan speed needs to be increased by 20% and the high-temperature stage compressor frequency needs to be increased by 5 Hz, so that the condensing temperature is maintained at 40 DEG C, and the whole refrigeration system performance is stable. After air pressure compensation, the system refrigeration capacity can still reach more than 95% of the rated value under the low air pressure environment of 75 kPa.

[0032] Through the implementation of step 1, an ultralow-temperature environment data set is obtained, which includes: the temperature in the box body is -65 DEG C, 8-point temperature distribution data, three-stage compressor operating frequency, each stage evaporation pressure and condensing pressure, system energy efficiency ratio 0.85, environmental air pressure data, air pressure compensation parameters and other key operating parameters, which provide basic data support for the subsequent construction of the phase change energy storage system.

[0033] Step 2, based on the ultralow-temperature environment data set, the energy storage layer is constructed and intermittently operated, and the optimized ultralow-temperature environment data set is generated; Specifically, the following steps are included: Step 2.1, based on the phase change requirement of the ultralow-temperature zone, a nano-nucleating agent reinforced composite phase change material is prepared; Based on the requirement that the phase change material needs to undergo phase change at minus 67 degrees Celsius to maintain the temperature stability of the box, a low-temperature eutectic mixture is used as the base material and nano-nucleating agent and thermal conductivity enhancer are added to obtain a performance-optimized composite phase change material. The reason for choosing minus 67 degrees Celsius as the phase change temperature point is that this temperature is 2 degrees Celsius lower than the target temperature of minus 65 degrees Celsius, ensuring that the phase change material can effectively absorb and release heat when the temperature of the box fluctuates, while avoiding the slow response of the phase change material caused by too low temperature. In one embodiment, a eutectic mixture of n-pentane and n-hexane is selected as the phase change base material, n-pentane has a molecular formula of C5H12, a melting point of minus 129.7 degrees Celsius, a boiling point of 36.1 degrees Celsius, and a latent heat of fusion of 116 kilojoules per kilogram, n-hexane has a molecular formula of C6H14, a melting point of minus 95 degrees Celsius, a boiling point of 68.7 degrees Celsius, and a latent heat of fusion of 152 kilojoules per kilogram, after mixing the two alkanes in a mass ratio of 6 to 4, according to the eutectic phase diagram theory, the eutectic point temperature of the mixture is about minus 67 degrees Celsius, and the mixed phase change latent heat is about 130 kilojoules per kilogram. To improve the phase change latent heat, 20% by mass of cyclohexane is further added as the third component, cyclohexane has a molecular formula of C6H12, a melting point of 6.5 degrees Celsius, and a latent heat of fusion of 32 kilojoules per kilogram, the ternary mixture is adjusted to have a stable phase change temperature of minus 67 degrees Celsius, and the phase change latent heat is increased to 165 kilojoules per kilogram.

[0034] To reduce the supercooling degree of the phase change process, nano-graphene oxide is added as a nucleating agent in the mixed alkanes, the nano-graphene oxide has a particle size range of 50 nanometers to 80 nanometers and a specific surface area of 800 square meters per gram, and the addition amount is 0.3% by mass, the nano-particles provide heterogeneous nucleation sites during the phase change process, reduce the supercooling degree required for nucleation, and reduce the supercooling degree of the phase change material from 8 degrees Celsius to 10 degrees Celsius when no nano-particles are added to about 2.5 degrees Celsius. The addition amount of 0.3% is based on the following considerations: if the addition amount is too low, there are not enough nucleation sites, and the supercooling degree reduction effect is limited; if the addition amount is too high, the nano-particles are easy to agglomerate, which affects the phase change performance. Through orthogonal experiments, it is determined that 0.3% is the optimal addition amount.

[0035] To enhance the thermal conductivity of the phase change material, expanded graphite is added as a thermal conductivity enhancer, the expanded graphite is prepared from natural flake graphite through acid intercalation and high-temperature expansion, has a volume expansion ratio of more than 200 times, and a thermal conductivity of 5 W / (m·K), and the addition amount is 15% by mass, the expanded graphite forms a three-dimensional thermal conductivity network in the phase change material, and the thermal conductivity of the composite phase change material is increased from 0.2 W / (m·K) of pure alkanes to 6.8 W / (m·K). The addition amount of 15% is the result of balancing the thermal conductivity and the phase change latent heat: if the addition amount is less than 10%, the thermal conductivity is not significantly improved, and if the addition amount is more than 20%, the phase change latent heat is significantly reduced and the material flowability is poor. Through thermal conductivity testing and phase change enthalpy testing, it is determined that 15% is the optimal addition amount.

[0036] The preparation process of the composite phase change material includes: first, mixing n-pentane, n-hexane and cyclohexane according to a mass ratio of 6:4:2, since these alkanes are in liquid state at room temperature, directly mixing them uniformly in a closed container; dispersing nano-oxidized graphene in anhydrous ethanol to form a suspension, treating by ultrasonic dispersion, ultrasonic frequency 20 kHz, power 400 W, dispersion time 30 minutes, so as to fully disperse the nanoparticles; slowly adding the nano-oxidized graphene suspension into the alkane mixture, mechanically stirring at 25 degrees Celsius for 2 hours at a speed of 500 revolutions per minute, so as to uniformly distribute the nanoparticles; then adding expanded graphite and continuing to stir for 1 hour; drying in a vacuum drying oven at 25 degrees Celsius under a vacuum degree of 10 Pa for 12 hours to remove the residual ethanol solvent, obtaining the nano-nucleating agent reinforced composite phase change material. Since the low-carbon alkanes have strong volatility, the entire preparation process is carried out in a closed system to prevent the change of the proportion caused by the volatilization of the components.

[0037] Step 2.2, packaging the composite phase change material and installing it on the inner wall of the box to obtain a phase change energy storage layer; Based on the requirement that the prepared composite phase change material needs to be packaged and fixed on the inner wall of the box, an aluminum alloy honeycomb panel is used as a packaging container to form an energy storage layer structure, and a phase change energy storage layer installed on the inner wall of the box is obtained. In an embodiment, the packaging container is an aluminum alloy honeycomb panel, the honeycomb core is a 3003 aluminum alloy foil with a thickness of 0.08 mm, the honeycomb aperture is 6 mm, the honeycomb height is 30 mm, the faceplate is an aluminum alloy plate with a thickness of 1 mm, and the faceplate and the honeycomb core are bonded by epoxy resin to form a closed honeycomb cavity. Since the low-carbon alkanes are in liquid state at room temperature, the composite phase change material is injected into the honeycomb cavity at room temperature through the injection port, the filling rate of each honeycomb cavity is 85%, and 15% of space is reserved as compensation space for volume change and thermal expansion during the phase change. After the injection is completed, the injection port is sealed with aluminum foil tape, and an epoxy resin sealant is coated on the outer layer to form a double-sealing structure to prevent the volatilization and leakage of low-boiling-point alkanes.

[0038] To further enhance the heat transfer inside the honeycomb panel, a copper heat-conducting fin with a thickness of 0.5 mm, a width of 5 mm and a height of 28 mm is inserted into the honeycomb cavity every 50 mm, the fin is welded and fixed to the honeycomb faceplate, the copper fin has a heat conductivity coefficient of 398 W / (m·K), and a fast heat conduction channel is established between the phase change material and the honeycomb faceplate. The size of the packaged phase change energy storage honeycomb panel is 800 mm long, 600 mm wide and 30 mm thick, and the weight of a single panel is about 8 kg, of which the mass of the composite phase change material is about 5 kg.

[0039] Phase change energy storage (PCE) honeycomb panels are installed on the inner wall of the enclosure. The enclosure has internal dimensions of 1200 mm (length), 800 mm (width), and 800 mm (height), with an effective volume of 0.768 cubic meters. Two PCE panels are installed on each of the left and right inner walls, one on each of the front and rear inner walls, and two on the bottom, for a total of eight PCE panels. The total mass of the energy storage material is 40 kg. Based on a latent heat of phase change of 165 kJ / kg, the total energy storage capacity is 6600 kJ. The PCE panels are fixed to the inner wall of the enclosure using stainless steel clips, with a 3 mm gap between the clips and the wall. This gap is filled with aerogel felt as an additional insulation layer to prevent thermal bridging between the energy storage panels and the enclosure wall.

[0040] Step 2.3: Based on the phase change energy storage layer, the intermittent operation of the refrigeration system is controlled to obtain a stable ultra-low temperature environment; Based on the characteristic of the phase change energy storage layer installed on the inner wall of the chamber in step 2.2, which can release cold energy when the refrigeration system is shut down, an intermittent operation control strategy for the refrigeration system is adopted to obtain a stable ultra-low temperature environment with temperature fluctuations controlled within ±0.5 degrees Celsius and reduce energy consumption. In one embodiment, the operation of the refrigeration system is divided into a cooling stage and an insulation stage, which alternate. In the cooling stage, the three-stage compressor operates at a set frequency, and the evaporator inside the chamber continuously absorbs heat from the chamber. The chamber temperature drops from -64.5 degrees Celsius to -65.5 degrees Celsius, a temperature drop of 1 degree Celsius. During this process, the temperature of the phase change energy storage layer decreases synchronously. When the temperature of the energy storage layer drops to -68 degrees Celsius, the composite phase change material undergoes a liquid-solid phase change, changing from a liquid to a solid state. During the phase change, 165 kJ of latent heat of phase change is released per kilogram, and a total energy storage of 6600 kJ is stored in the energy storage layer. The cooling stage lasts for about 10 minutes.

[0041] After the cooling stage ends, the insulation stage begins. The control system shuts down the three-stage compressor, and the cooling system stops operating. The temperature inside the chamber gradually rises due to the intrusion of external heat. When the temperature reaches -64.5 degrees Celsius, the temperature of the phase change energy storage layer rises to -66 degrees Celsius. The composite phase change material undergoes a reverse phase change, transforming from a solid to a liquid state. During the phase change, it absorbs heat from the environment, and the latent heat of phase change is released to maintain the chamber temperature. Since the rate of external heat intrusion into the chamber is approximately 130 watts, the phase change energy storage layer can theoretically maintain a stable temperature for approximately 850 minutes (6600 kJ divided by 130 watts). However, considering the requirements for temperature control accuracy and the limitations of the heat transfer rate of the phase change material, the actual insulation stage duration is set to 15 minutes. At this time, the chamber temperature rises to -64.5 degrees Celsius, and the temperature fluctuation range is still within ±0.5 degrees Celsius.

[0042] After the insulation phase ends, the refrigeration system is restarted to begin the next refrigeration phase, and this cycle repeats. Through intermittent operation control, the refrigeration system's operating time accounts for 40% of the total time (10 minutes divided by 25 minutes). Compared to continuous operation, intermittent operation reduces the refrigeration system's operating time by 60%. However, considering the transient energy consumption during startup, the actual energy saving rate is approximately 50%. During the 48-hour transportation process, the total energy consumption decreased from 134.4 kWh in continuous operation to 67.2 kWh in intermittent operation.

[0043] In another embodiment, since the performance of phase change materials may degrade after multiple phase change cycles, microencapsulation technology can be used to improve the cycle stability of phase change materials, with the aim of extending the service life of the phase change energy storage system. Preferably, the composite phase change material prepared in step 2.1 is made into microcapsules using in-situ polymerization with melamine-formaldehyde resin as the wall material. The composite phase change material is dispersed in an aqueous solution containing an emulsifier to form an emulsion. Melamine and formaldehyde are added and subjected to a condensation reaction under acidic conditions. The resin polymerizes on the surface of the phase change material droplets to form a shell. The microcapsule particle size ranges from 10 micrometers to 50 micrometers, and the wall thickness ranges from 1 micrometer to 3 micrometers. The wall material encapsulates and isolates the phase change material to prevent component stratification or leakage during multiple melting and solidification processes.

[0044] By implementing step 2, an optimized ultra-low temperature environment dataset is obtained. This dataset includes key parameters such as: temperature fluctuation range ±0.5 degrees Celsius, phase change material formulation parameters, phase change temperature -67 degrees Celsius, latent heat of phase change 165 kJ / kg, total energy storage 6600 kJ, cooling stage time 10 minutes, heat preservation stage time 15 minutes, intermittent operation ratio 40%, and energy saving rate 50%. This provides a stable temperature environment foundation for the subsequent construction of the vibration reduction system.

[0045] Step 3: Based on the optimized ultra-low temperature environment dataset, magnetorheological vibration reduction is performed and combined with a collaborative control strategy to obtain a vibration reduction and protection control scheme; Specifically, the following steps are included: Step 3.1: Install magnetorheological vibration dampers at the bottom of the box to obtain an adjustable damping vibration support system; Based on the requirement that the enclosure needs to isolate external vibrations, a magnetorheological damper is used as the connecting element between the enclosure and the external support frame, resulting in a vibration damping support system with an adjustable damping coefficient in real time. In one embodiment, one set of magnetorheological vibration dampers is installed at each of the four corners of the bottom of the housing. The damper structure includes an outer cylinder, a piston rod, a piston head, a magnetorheological fluid chamber, and an excitation coil. The outer cylinder is made of non-magnetic stainless steel, with an inner diameter of 40 mm, a wall thickness of 3 mm, and a length of 150 mm. The piston rod has a diameter of 20 mm and is made of high-strength alloy steel with a chrome-plated surface to improve wear resistance. The piston head has a diameter of 38 mm and a 1 mm gap between it and the inner wall of the outer cylinder. The piston head has an annular groove inside for installing the excitation coil. The coil has 200 turns and a wire diameter of 0.5 mm. The excitation current can be adjusted from 0 amperes to 2 amperes. The magnetorheological fluid chamber is filled with magnetorheological fluid, which consists of micron-sized carbonyl iron powder, silicone oil carrier, dispersant, and stabilizer. The iron powder volume fraction is 35%. The apparent viscosity is 0.1 Pa·s under zero magnetic field and can reach 50 Pa·s under saturated magnetic field, with an adjustment range of 500 times.

[0046] The working principle of the magnetorheological damper is as follows: when the piston rod moves relative to the outer cylinder under external vibration, the magnetorheological fluid is squeezed through the annular gap between the piston head and the inner wall of the outer cylinder. Under zero magnetic field conditions, the magnetorheological fluid exhibits low-viscosity Newtonian fluid characteristics, with relatively small damping force. When the excitation coil is energized to generate a magnetic field, the iron powder particles in the magnetorheological fluid align along the magnetic field lines to form a chain-like structure, and the fluid rheological properties transform into Bingham plasticity, with a sharp increase in apparent viscosity and increased damping force. The damping coefficient can be changed in real time by adjusting the excitation current, with a response time of less than 10 milliseconds. Four sets of magnetorheological dampers isolate the housing from the external support frame, which is fixed to an aviation transport pallet. The housing is suspended from the support frame by the dampers, forming a one-degree-of-freedom vibration isolation system.

[0047] Step 3.2: Based on the vibration signal collected by the triaxial accelerometer and the optimal damping force calculated, the excitation current control command of the vibration damper is obtained; Magnetorheological vibration dampers require adjustment of the damping coefficient based on real-time vibration conditions. A triaxial accelerometer is used to collect vibration signals, and a semi-active control algorithm is employed to calculate the optimal damping force, thereby obtaining excitation current control commands for each damper. In one embodiment, a triaxial accelerometer is installed at the center of the base plate of the enclosure. The sensor is manufactured using MEMS technology, with a measurement range of ±5 times gravitational acceleration, a resolution of 0.001 times gravitational acceleration, and a sampling frequency of 1000 Hz. The sensor outputs acceleration signals in three directions (X, Y, and Z axes), which are transmitted to the vibration damping controller via data lines.

[0048] The vibration reduction controller employs a semi-active control strategy based on a ceiling damping control algorithm. This algorithm utilizes the concept of a virtual ceiling damper, increasing the damping force to suppress vibration when the vibration velocity and acceleration of the enclosure are in the same direction, and decreasing the damping force to avoid amplifying vibration when they are in opposite directions, thus achieving a near-optimal vibration reduction effect. The controller first performs numerical integration on the acceleration signal to obtain the vibration velocity signal. The integration algorithm uses the trapezoidal rule, where the current velocity equals the previous velocity plus the average acceleration of the current and previous moments multiplied by the time interval, which is 1 millisecond.

[0049] The desired damping force is calculated based on the ceiling damping control law. The desired damping force is equal to the negative ceiling damping coefficient multiplied by the vibration velocity. The ceiling damping coefficient is set to 3000 N / m. Since the magnetorheological damper can only provide positive damping force and cannot provide negative driving force, when the calculated desired damping force is negative, the desired damping force is set to 0.

[0050] The required apparent viscosity of the magnetorheological fluid is calculated based on the desired damping force using the vibration damper's mechanical model. The damping force is equal to 12 times pi multiplied by the apparent viscosity of the magnetorheological fluid multiplied by the cube of the piston radius multiplied by the piston length divided by the annular gap. According to the magnetorheological effect model of the magnetorheological fluid, the relationship between the apparent viscosity and the magnetic field strength is a hyperbolic tangent function. The required magnetic field strength is obtained by looking up a table or iterative calculation. The magnetic field strength is proportional to the excitation current, and finally, the excitation current control command is obtained, with a current range of 0 amperes to 2 amperes.

[0051] The controller sends current commands to the current drive module, which uses pulse width modulation (PWM) to control the excitation coil current with a switching frequency of 20 kHz and a current control accuracy of 0.01 amperes. Four sets of magnetorheological dampers are controlled independently, with each damper adjusting its damping force according to the vibration state at its location to suppress the six-degree-of-freedom vibration of the enclosure.

[0052] Step 3.3: Based on the vibration reduction effect combined with the vibration-temperature synergistic control strategy, a protective effect of reduced tissue damage rate is obtained; Magnetorheological vibration damping systems can effectively reduce vibration transmission under normal vibration conditions, but under extreme vibration conditions, temperature regulation is still needed to enhance the tissue strength of aquatic products, employing a vibration-temperature coordinated control strategy. In one embodiment, a vibration intensity assessment module is integrated into the vibration damping controller. This module calculates the root mean square (RMS) value of vibration acceleration in real time. The RMS value is equal to the square root of the average of the sum of squares of acceleration within a time window. The time window is set to 1 second containing 1000 sampling points. The vibration RMS value reflects the average intensity of the vibration. A vibration intensity threshold of 1.5 times the gravitational acceleration is set. When the detected vibration RMS value exceeds this threshold, it is determined to be a severe vibration state. At this time, it is difficult for the vibration damper alone to control the peak vibration acceleration below 1.2 times the gravitational acceleration, thus activating the vibration-temperature coordinated control strategy.

[0053] The coordinated control strategy involves sending a cooling command to the temperature control system in step 2, lowering the chamber temperature setpoint from -65°C to -67°C (a decrease of 2°C). The refrigeration system increases its cooling power, and the frequencies of the three compressor stages are increased by 5 Hz. After approximately 10 minutes, the chamber temperature reaches -67°C. At this lower temperature, the molecular thermal motion of the bluefin tuna muscle tissue decreases, and the elastic modulus increases. At -65°C, the elastic modulus of tuna muscle is approximately 500 MPa, increasing to 550 MPa at -67°C (an increase of 10%). This higher elastic modulus reduces the deformation of the muscle tissue under vibration stress, lowering the risk of tissue damage. Simultaneously, the lower temperature stabilizes the myoglobin molecular structure, reducing the rate of vibration-induced oxidation.

[0054] After implementing the collaborative control strategy, under external vibration input conditions of 3 times the gravitational acceleration, the vibration transmitted to the tank was reduced to 1.8 times the gravitational acceleration through the magnetorheological vibration reduction system, achieving a vibration reduction rate of 40%. Combined with the enhanced tissue strength due to temperature reduction, the elastic modulus of the tuna muscle tissue increased by 10%. According to the theory of materials mechanics, under the same vibration acceleration, the internal stress of the tissue is inversely proportional to the elastic modulus. Therefore, the internal stress of the tissue was reduced by approximately 9%. Considering the combined effect of vibration reduction and tissue strength improvement, the actual damage risk borne by the aquatic product tissue was equivalent to the damage risk under vibration conditions of 1.6 times the gravitational acceleration. Experimental verification showed that the tissue damage rate under this condition was lower than the damage rate under direct action of 1.2 times the gravitational acceleration, meeting the safety threshold requirements.

[0055] By implementing step 3, a vibration reduction and protection control scheme is obtained. This scheme includes key control parameters such as the structural parameters of the magnetorheological vibration damper, the magnetorheological fluid formula, the ceiling damping control algorithm parameters, the vibration-temperature coordinated control strategy, the vibration reduction rate of 40%, and the vibration acceleration control target of less than 1.2 times the gravitational acceleration. This provides a foundation for vibration reduction and protection for the stable operation of the subsequent quality monitoring system.

[0056] Step 4: Based on the aquatic products in the vibration reduction and protection control scheme, perform defrosting-coordinated multispectral sensing and quality prediction to generate a quality monitoring dataset; Specifically, the following steps are included: Step 4.1: Install the multispectral sensor and configure the defrosting device inside the enclosure; The -65°C ultra-low temperature environment inside the enclosure can cause frost formation on the optical probe surface, affecting spectral acquisition. A stable monitoring hardware system is achieved by combining a spectral sensor with a miniature heating and defrosting device. In one embodiment, a spectral monitoring component is installed at the top front of the enclosure. This component includes a near-infrared spectrometer, a visible light spectrometer, an LED light source, and a heating and defrosting device. The near-infrared spectrometer employs a miniature grating beam splitting structure, with a wavelength range of 900 nm to 1700 nm and an optical resolution of 5 nm. Its detector is an indium gallium arsenide (InGaAs) linear array with 512 pixels, a sampling interval of 2 nm, and an adjustable integration time ranging from 1 ms to 1000 ms. The visible light spectrometer employs a linear variable filter structure, with a wavelength range of 400 nm to 700 nm and an optical resolution of 3 nm. Its detector is a silicon-based CMOS linear array with 256 pixels.

[0057] The LED light source uses a broadband white LED with an emission wavelength range of 400 nm to 1800 nm and a power of 1 watt. The LED chip is packaged on an aluminum substrate and contacts the heat sink via thermal grease. The light source uses a pulsed drive mode with a 10% duty cycle to reduce heat generation. The optical probe uses a coaxial illumination and collection structure. The light emitted by the LED light source is guided to the front end of the probe through an optical fiber and illuminates the surface of the tuna at a 45-degree angle. The diffused light is collected by the collecting optical fiber at another 45-degree angle to prevent specular reflection from entering the detector. The collecting optical fiber then transmits the light to the spectrometer for spectroscopic detection.

[0058] A miniature heating defrosting device is arranged in a ring around the optical probe. The heating element is a thin-film resistance heating element with a power of 5 watts and dimensions of 30 mm long, 20 mm wide, and 0.2 mm thick. The heating element is attached to the optical probe housing with thermally conductive adhesive. A temperature sensor, using an NTC thermistor, is attached to the probe surface, with a temperature measurement range of -50°C to 50°C. The heating defrosting uses a pulse heating mode with a heating cycle of 60 seconds. During the first 5 seconds of each cycle, the heating element is powered on, controlling the probe surface temperature to rise to -20°C. At this temperature, the frost layer within a 3 mm range on the probe surface sublimates, reducing the frost layer thickness from 0.5 mm before heating to below 0.05 mm. Spectral acquisition is performed from the 6th to the 8th second after heating, when the frost layer has minimal impact. After acquisition, the heating element is de-powered, and the probe surface temperature gradually decreases to the ambient temperature of -65°C within 55 seconds. The average power consumption of the heating defrosting device is 5 watts multiplied by 5 seconds divided by 60 seconds, which equals 0.42 watts, having a minimal impact on overall energy consumption.

[0059] Step 4.2: Collect spectral data of tuna surface; Based on the multispectral sensor, the reflectance spectrum of the tuna surface is collected during the defrosting window. A characteristic wavelength extraction method is used to obtain spectral characteristic parameters for quality assessment. In one embodiment, the spectral acquisition process is as follows: at the 6th second of each 60-second heating defrosting cycle, the control system simultaneously triggers the LED light source, near-infrared spectrometer, and visible light spectrometer. The LED light source is turned on, with an integration time of 50 milliseconds. Light shines onto the tuna surface, and diffuse reflected light is collected by the collecting fiber and transmitted to the two spectrometers. The near-infrared spectrometer and visible light spectrometer simultaneously acquire spectral data. Each spectrometer's acquisition time is 2 seconds, including 40 integrations to improve the signal-to-noise ratio. After acquisition, the LED light source is turned off, and the spectral data is transmitted to the quality analysis controller via an RS-485 bus.

[0060] Visible light spectroscopy is used to detect the degree of myoglobin oxidation. Myoglobin is the main pigment protein in tuna muscle. Fresh myoglobin is bright red and has characteristic absorption peaks at 410 nm and 525 nm. When myoglobin is oxidized to methemoglobin, its color turns brown, the absorption peak at 410 nm weakens, and the absorption peaks at 540 nm and 630 nm strengthen. The reflectance at four characteristic wavelengths (410 nm, 525 nm, 540 nm, and 630 nm) is extracted from the visible light spectral data and denoted as R410, R525, R540, and R630. The myoglobin oxidation index (MbOI) is calculated as R630 divided by R525. This index is positively correlated with the degree of myoglobin oxidation. Fresh tuna has an MbOI less than 0.5, mildly oxidized MbOI is between 0.5 and 1.0, moderately oxidized MbOI is between 1.0 and 1.5, and severely oxidized MbOI is greater than 1.5.

[0061] Near-infrared spectroscopy is used to detect the degree of lipid peroxidation and water and protein content. In the near-infrared band, 980 nm corresponds to the second-order overtone absorption of the OH bond in water molecules, 1210 nm corresponds to the second-order overtone absorption of the CH bond in lipids, 1450 nm corresponds to the combined overtone absorption of the NH and OH bonds in proteins, and 1730 nm corresponds to the first-order overtone absorption of the C=O group in lipids. Lipid peroxidation leads to an increase in carbonyl content, which enhances the absorption peak at 1730 nm. The absorbance of the four characteristic wavelengths (980 nm, 1210 nm, 1450 nm, and 1730 nm) is extracted from the near-infrared spectral data and denoted as A980, A1210, A1450, and A1730. The absorbance is equal to the negative logarithm of the reflectance.

[0062] Step 4.3: Based on the spectral characteristic parameters, the degree of myoglobin oxidation and the lipid peroxidation value are calculated using a quality prediction model. Based on visible and near-infrared spectral characteristic parameters, a partial least squares regression (PLSR) prediction model was used to obtain quantitative results for the percentage of myoglobin oxidation and lipid peroxidation value. In one embodiment, the quality prediction model needs to be established with a large number of calibration samples before use. The calibration process is as follows: 120 tuna samples with different shelf life from 0 to 72 hours were collected, with sampling every 6 hours, and 10 parallel samples were collected each time. Spectroscopic measurements and chemical analyses were performed on each sample simultaneously. The spectral measurements yielded 8 characteristic parameters, namely R410, R525, R540, R630, A980, A1210, A1450, and A1730. The chemical analysis measured the degree of myoglobin oxidation and lipid peroxidation value as reference values. The degree of myoglobin oxidation was determined by spectrophotometry, and the percentage of metmyoglobin in total myoglobin was calculated. The lipid peroxidation value was determined by the thiobarbituric acid method (TBARS). The results were expressed as malondialdehyde (MDA) equivalents in milligrams per kilogram.

[0063] Using spectral characteristic parameters of 120 samples as independent variables (X matrix) and myoglobin oxidation degree and lipid peroxidation value as dependent variables (Y matrix), a prediction model was established using partial least squares regression. PLSR established a linear regression relationship by extracting principal components from the X and Y matrices and maximizing their covariance. The number of principal components was determined to be 5 through cross-validation. The model form is Y equal to X multiplied by the regression coefficient matrix B plus a constant term. The regression coefficient matrix B is 8 rows and 2 columns, corresponding to the regression coefficients from the 8 spectral characteristic parameters to the 2 quality indicators.

[0064] After model establishment, validation was performed using leave-one-out cross-validation. The coefficient of determination (R²) for the myoglobin oxidation prediction model was 0.93, and the root mean square error (RMSEP) was 1.8 percentage points. The coefficient of determination (R²) for the lipid peroxidation prediction model was 0.95, and the RMSEP was 0.08 mg / kg. The prediction accuracy met the requirements for real-time monitoring. During actual transportation, the quality analysis controller inputs spectral characteristic parameters into the prediction model to calculate the current myoglobin oxidation level and lipid peroxidation value. The data is updated every 60 seconds to form a time series of quality indicators. The quality data is then transmitted to a cloud monitoring platform via a wireless communication module, allowing users to view the freshness status of the tuna in real time.

[0065] In another embodiment, since the optical characteristics of tuna vary among different species or parts, transfer learning can be used to improve the model's generalization ability. The aim is to make the prediction model applicable to a wider range of sample types. Preferably, during the calibration phase, samples of multiple tuna species, including yellowfin tuna, bigeye tuna, and bluefin tuna, as well as samples of different parts of the same species, including red meat from the back, fatty belly, and medium fat, are collected, totaling 300 calibration samples. A general prediction model is then established. Then, for a specific application scenario, a small number of samples from that scenario are collected as transfer samples. The regression coefficient matrix is ​​adjusted through a model adaptive algorithm, so that the model can maintain its generality while optimizing the prediction accuracy for specific samples. After transfer learning, the prediction determination coefficient of the model for specific samples can be increased to R-squared or higher than 0.97.

[0066] Step 4 yields a quality monitoring dataset containing key monitoring data such as near-infrared spectral data, visible light spectral data, eight spectral characteristic parameters, myoglobin oxidation level, lipid peroxidation value, defrosting control parameters, and a data update frequency of every 60 seconds. This dataset provides real-time quality feedback for the subsequent implementation of multi-parameter collaborative temperature control strategies.

[0067] Step 5: Based on the quality monitoring dataset, the optimized ultra-low temperature environment dataset, and the ultra-low temperature environment dataset, a three-layer hierarchical control strategy is adopted to obtain a multi-parameter collaborative control scheme. Specifically, the following steps are included: Step 5.1: Implement a fast-response temperature control layer based on an FPGA chip; The temperature inside the enclosure fluctuates rapidly due to various disturbances, requiring timely correction. A simplified fuzzy PID control algorithm running on an FPGA chip is employed to achieve a fast response capability with a control cycle of 100 milliseconds. In one embodiment, the inputs to the fast response control layer are the deviations between the current temperature value measured by eight temperature sensors inside the enclosure and the set temperature value. The average temperature deviation is obtained by averaging the eight deviation values, and the rate of change of the average temperature deviation is obtained by subtracting the deviation from the previous cycle from the current deviation and then dividing by the time interval. The temperature deviation and the rate of change of the deviation are used as the two input variables of the fuzzy controller.

[0068] The fuzzy universe of discourse for temperature deviation is divided into three fuzzy subsets: negative, zero, and positive. The corresponding membership functions are trigonometric functions. The negative subset corresponds to temperature deviations less than 0.3 degrees Celsius, the zero subset corresponds to temperature deviations between -0.3 and +0.3 degrees Celsius, and the positive subset corresponds to temperature deviations greater than 0.3 degrees Celsius. The fuzzy universe of discourse for the rate of change of deviation is also divided into three fuzzy subsets: negative, zero, and positive. The output variable is the compressor frequency increment, and its fuzzy universe of discourse is divided into five fuzzy subsets: negative large, negative small, zero, positive small, and positive large.

[0069] A simplified fuzzy rule table is established, consisting of 7 rules: Rule 1: If the temperature deviation is negative and the rate of change of the deviation is negative, then a large positive frequency increment indicates that the temperature is too low and will continue to decrease, requiring a significant reduction in cooling power; Rule 2: If the temperature deviation is negative and the rate of change of the deviation is 0, then a small positive frequency increment; Rule 3: If the temperature deviation is negative and the rate of change of the deviation is positive, then a zero frequency increment; Rule 4: If the temperature deviation is 0, then a zero frequency increment; Rule 5: If the temperature deviation is positive and the rate of change of the deviation is negative, then a zero frequency increment; Rule 6: If the temperature deviation is positive and the rate of change of the deviation is 0, then a small negative frequency increment; Rule 7: If the temperature deviation is positive and the rate of change of the deviation is positive, then a large negative frequency increment indicates that the temperature is too high and will continue to rise, requiring a significant increase in cooling power.

[0070] Fuzzy inference employs the Mamdani minimum inference method, while defuzzification uses the centroid method to calculate the precise output value. The field-programmable gate array (FPGA) chip is a Xilinx 7th generation series, with 10,000 logic cells and an operating frequency of 100 MHz. The fuzzy control algorithm is implemented in the FPGA using a hardware description language and a pipelined architecture. Fuzzification, rule-based inference, and defuzzification each occupy one clock cycle, resulting in a total delay of 3 clock cycles (30 nanoseconds). Including the sampling delay from analog-to-digital conversion and the output delay from digital-to-analog conversion, the total control delay is less than 1 microsecond. The control cycle is set to 100 milliseconds, meaning the control output is updated 10 times per second. The output of the fast-response control layer is the frequency increment of the three-stage compressor, with an increment range of ±5 Hz. This frequency increment is superimposed on the current frequency setpoint to obtain a new frequency command, which is then sent to the compressor inverter driver via a pulse-width modulation (PWM) signal.

[0071] Step 5.2: Implement the optimized decision-making temperature control layer based on the ARM processor; The fast response layer, which only considers temperature deviation, cannot handle multi-parameter collaborative optimization. Therefore, a model predictive control algorithm running on a high-reduction instruction set processor is employed to obtain the optimal control sequence that comprehensively considers temperature, air pressure, and quality indicators. In one embodiment, the inputs to the optimization decision control layer include 11 state variables: 8-point temperature distribution within the chamber, ambient air pressure, myoglobin oxidation level, and lipid peroxidation value. The outputs are the three-stage compressor frequency setpoint and the electronic expansion valve opening setpoint. A system prediction model is established, adopting a state-space form. The state equation is: the state at the next moment equals the state transition matrix multiplied by the current state, plus the control input matrix multiplied by the current control quantity, plus the disturbance input matrix multiplied by the current disturbance. The state is an 11-dimensional state vector including 8 temperature, air pressure, and 2 quality indicators. The control quantity is a 5-dimensional control vector including 3 compressor frequencies and 2 expansion valve openings. The expansion valves of the low temperature stage and the medium temperature stage are independently controlled, and the opening of the high temperature stage expansion valve is coupled with the high temperature stage compressor frequency. The disturbance is a disturbance vector including the external ambient temperature and heat load. The matrix parameters such as the state transition matrix, control input matrix, and disturbance input matrix are obtained through system identification methods and fitted based on historical operating data using the least squares method.

[0072] The objective function of the model predictive control algorithm is the sum of performance indices from the current time to the end of the prediction time domain, where the prediction time domain is set to 10 minutes, i.e., 10 sampling periods. The objective function includes three terms: the first term is the weighted sum of squares of temperature tracking error, with a weight coefficient of 10; the second term is a quality deterioration penalty term, which applies a large weight penalty when the myoglobin oxidation rate or lipid peroxidation value exceeds a threshold, with a weight coefficient of 100; and the third term is a penalty term for changes in control input, which suppresses frequent and large changes in control input, with a weight coefficient of 1. Constraints include a compressor frequency range of 30 to 90 Hz, an expansion valve opening range of 10% to 100%, and a temperature range of -68 to -64 degrees Celsius.

[0073] Model predictive control solves a constrained quadratic programming problem in each control cycle to obtain the optimal control sequence at multiple moments within the future control time domain. The control time domain is set to 2 minutes, i.e., 2 sampling cycles. Only the first control variable is executed, and the optimization problem is resolved based on the new measurements in the next cycle, forming a rolling optimization. The quadratic programming solution uses the interior-point method, and the 7th generation architecture of the RISC processor is used with a clock frequency of 1.2 GHz. The single optimization solution time is approximately 500 milliseconds, and the control cycle is set to 1 minute to meet the real-time requirements.

[0074] The optimized decision-making layer also implements a collaborative control strategy. When the rate of myoglobin oxidation is detected, i.e., the increase in oxidation degree divided by the time interval is greater than the threshold of 0.5% per hour, it is determined that the oxidation reaction is accelerating. The temperature setpoint is immediately reduced from -65 degrees Celsius to -67 degrees Celsius, and the nitrogen replacement system in the chamber is activated at the same time. The nitrogen replacement system includes a nitrogen storage tank and a solenoid valve. The nitrogen storage tank has a volume of 10 liters, a pressure of 15 MPa, and stores 1.8 kg of nitrogen. After the pressure is reduced to 0.2 MPa through the pressure reducing valve, the nitrogen is injected into the chamber through the solenoid valve. The chamber has a volume of 0.768 cubic meters, and under standard conditions, a nitrogen volume of 0.73 cubic meters and a mass of 0.9 kg are required. After the nitrogen is injected, the oxygen concentration in the chamber decreases from 21% to below 5%. The low-oxygen environment inhibits the myoglobin oxidation reaction and the lipid auto-oxidation reaction.

[0075] When the ambient air pressure is detected to drop from 101 kPa to below 85 kPa, the aircraft is determined to have entered the cruise phase. Due to the decrease in air pressure, the heat dissipation capacity of the high-temperature stage condenser decreases. The optimization algorithm automatically increases the frequency of the high-temperature stage compressor by 5 Hz and increases the speed of the condenser fan by 20% to compensate for the impact of air pressure and maintain stable cooling capacity.

[0076] Step 5.3: Implement the parameter self-tuning control layer based on the cloud server; The Model Predictive Control (MMCC) algorithm involves multiple weight coefficients and model parameters that require continuous optimization based on actual performance. A deep reinforcement learning algorithm running on a cloud server is employed to obtain optimal control parameters updated after each transportation task. In one embodiment, the parameter self-tuning control layer uses a deep deterministic policy gradient algorithm, a reinforcement learning method based on an actuator evaluator architecture, suitable for control problems in continuous action spaces. The actuator network generates the control policy. Inputs include system states such as temperature, air pressure, and quality indicators; outputs are the weight coefficients of the MMCC algorithm. The actuator network structure is a three-layer fully connected neural network: 11 neurons in the input layer corresponding to 11 state variables, 64 neurons in the first hidden layer, 32 neurons in the second hidden layer, and 3 neurons in the output layer corresponding to 3 weight coefficients. The activation function is a modified linear unit (MRU). The evaluator network is used to evaluate the state-action value function. The inputs are the system state and the action output by the executor. The output is a value representing the long-term cumulative reward of performing the action in the current state. The evaluator network structure is a 4-layer fully connected network. The state input and action input are merged in the first hidden layer. The first hidden layer has 64 neurons, the second hidden layer has 64 neurons, the third hidden layer has 32 neurons, and the output layer has 1 neuron.

[0077] The training process is divided into two stages: offline pre-training and online optimization. In the offline pre-training stage, 1000 virtual transportation task data are first generated through simulation. Each task includes different environmental conditions (temperature 15°C to 35°C, air pressure 75 kPa to 101 kPa, vibration 0.2 to 3 times the gravitational acceleration) and load conditions (aquatic product weight 30 kg to 60 kg). Based on the physical model established in steps 1 to 4 and the experimental calibration data, the simulation model generates state trajectory, control output, and quality index change data. The simulation data is used to initialize the actuator network and evaluator network parameters, enabling the model to have basic control capabilities.

[0078] During the online optimization phase, edge devices record the entire status trajectory, control output, and quality index changes in each actual transportation task. After the task is completed, the data is uploaded to the cloud server. The cloud server divides one transportation task into multiple time steps, each time step being 5 minutes, and calculates the instant reward for each time step. The reward function is designed as follows: the reward equals the absolute value of the negative temperature deviation minus the quality deterioration penalty minus the energy consumption penalty. The quality deterioration penalty is a penalty of -100 when the degree of myoglobin oxidation exceeds 3% or the lipid peroxidation value exceeds 0.5 mg / kg. The energy consumption penalty is the compressor power multiplied by the unit electricity price coefficient of 0.01.

[0079] Transition samples consisting of state, action, reward, and next state are stored in an experience replay buffer with a capacity of 10,000 samples. A priority experience replay strategy is employed, prioritizing samples with larger temporal difference errors for training. During training, 64 samples are randomly selected from the buffer to form a batch. The target value is calculated using a temporal difference method, and the evaluator network parameters are updated to make the predicted value approximate the target value. Then, the executor network parameters are updated using a policy gradient method to maximize the value. The model is updated every 10 actual transportation task data iterations, and the updated model parameters are distributed to edge devices for continuous optimization.

[0080] By implementing step 5, a multi-parameter collaborative control strategy is obtained. This strategy includes key control parameters such as fast response control layer parameters, optimization decision control layer parameters, parameter self-tuning control layer parameters, and control objectives, providing an optimized control foundation for the efficient operation of the subsequent energy recovery system.

[0081] Step 6: Based on a multi-parameter collaborative control scheme, perform thermoelectric conversion to obtain an energy recovery and utilization scheme; Specifically, the following steps are included: Step 6.1: Install a thermoelectric conversion module at the outlet of the high-temperature stage condenser to obtain DC power output based on temperature difference; Based on the condition that the refrigerant temperature at the outlet of the high-temperature stage condenser is 40 degrees Celsius while the ambient temperature is 25 degrees Celsius, resulting in a 15-degree Celsius temperature difference, a bismuth telluride-based thermoelectric conversion module is used to obtain DC power output. In one embodiment, a heat exchange copper plate is installed on the outlet pipe of the high-temperature stage condenser. The copper plate has dimensions of 200 mm long, 100 mm wide, and 5 mm thick. A groove with a width of 12 mm and a depth of 3 mm is cut on the lower surface of the copper plate. The refrigerant pipe is embedded in the groove, and the gap is filled with thermally conductive silicone grease to ensure efficient heat transfer from the refrigerant to the copper plate. Forty thermoelectric conversion modules are evenly arranged on the upper surface of the copper plate. Each module has dimensions of 40 mm long, 40 mm wide, and 3.6 mm thick. The thermoelectric modules are made of bismuth telluride-based material and consist of alternating negative and positive thermoelectric arms connected in series by copper connecting pieces to form a thermocouple array. A single module contains 127 pairs of thermoelectric arms.

[0082] The hot end of the thermoelectric module is bonded to a copper plate via thermal grease, while the cold end is attached to a heat-dissipating aluminum plate. The aluminum plate measures 200 mm long, 100 mm wide, and 5 mm thick, with fins machined on its surface. The fins are 20 mm high and spaced 5 mm apart, allowing heat exchange with ambient air through natural convection. When the hot end temperature is 40 degrees Celsius and the cold end temperature is 28 degrees Celsius, the temperature difference is 12 degrees Celsius. The Seebeck coefficient of a single thermoelectric module is 0.2 volts per switch, and the output voltage equals the Seebeck coefficient multiplied by the temperature difference, which equals 2.4 volts. The internal resistance is 1 ohm. The output power is maximum when the external load resistance equals the internal resistance, and the maximum output power equals the square of the voltage divided by 4 times the internal resistance, which equals 1.44 watts. However, considering actual load matching and conversion efficiency, the actual output power of a single module is approximately 1.2 watts.

[0083] Forty thermoelectric modules are connected in series, with a total output voltage of 96 volts and a total output power of 48 watts. The DC voltage output from the thermoelectric modules fluctuates significantly. A DC-DC step-down converter steps down the 96 volts and stabilizes it to 28 volts with a conversion efficiency of 90%. The stabilized output power is 43 watts. This power is used to supply the control system, sensor network, data acquisition module, and communication module. The total power consumption of the control system is approximately 30 watts. The thermoelectric recovery energy can meet 60% of the control system's power requirements, reducing the amount of power drawn from the main power supply and improving the overall energy efficiency of the system.

[0084] Step 6.2: Based on changes in ambient temperature, intelligent air-cooled heat dissipation control is adopted to obtain stable thermoelectric output power; The cold-end temperature of the thermoelectric conversion module is significantly affected by ambient temperature. In high-temperature environments, the increased cold-end temperature leads to a decrease in temperature difference and a drop in output power. Forced air cooling is employed to improve cold-end heat exchange efficiency, resulting in stable output power while maintaining a large temperature difference under various ambient temperatures. In one embodiment, a small axial flow fan is installed beside the heat dissipation aluminum plate. The fan has a diameter of 60 mm, a power of 3 watts, and an airflow of 30 cubic meters per hour. The fan is driven by a brushless DC motor with a speed adjustment range of 1000 rpm to 5000 rpm, controlled by pulse width modulation. The fan blades employ a high-efficiency aerodynamic design, generating sufficient airflow even at low speeds.

[0085] The fan controller integrates an output power detection module and an intelligent control algorithm to monitor the output power of the thermoelectric module in real time. When the detected output power is below 35 watts, it is determined that the cold end temperature is too high, resulting in insufficient temperature difference. The controller starts the fan and adjusts the fan speed according to the degree of power deviation; the lower the output power, the higher the fan speed. When the output power is below 30 watts, the fan runs at its maximum speed of 5000 rpm. When the output power is between 30 watts and 35 watts, the fan speed is linearly adjusted between 3000 rpm and 5000 rpm. When the output power recovers to above 40 watts, it is determined that the temperature difference is sufficient. The controller reduces the fan speed or shuts down the fan to save energy. When the output power is between 40 watts and 45 watts, the fan runs at a low speed of 1000 rpm to 3000 rpm to maintain heat dissipation. When the output power exceeds 45 watts, the fan is completely shut down, relying on natural convection for heat dissipation.

[0086] Through intelligent air-cooling control, under standard conditions of 25 degrees Celsius ambient temperature, the cold end temperature is maintained at 28 degrees Celsius, with a temperature difference of 12 degrees Celsius. The thermoelectric module outputs 43 watts, and the fan is in the off state. Under high-temperature conditions of 35 degrees Celsius ambient temperature, without air-cooling, the cold end temperature will rise to 38 degrees Celsius, with a temperature difference of only 2 degrees Celsius, and the output power will drop to 7 watts. After activating intelligent air-cooling, the fan runs at 4000 rpm, and forced convection lowers the cold end temperature to 32 degrees Celsius, maintaining a temperature difference of 8 degrees Celsius. The thermoelectric module output power remains above 32 watts. After deducting the fan's power consumption of 3 watts, the net output power is still 29 watts, meeting the basic power requirements of the control system. Under low-temperature conditions of 15 degrees Celsius ambient temperature, the cold end temperature is 18 degrees Celsius, with a temperature difference of 22 degrees Celsius. The output power increases to 78 watts, and the fan remains off. The net output power of 78 watts provides sufficient power for the control system and charges the battery.

[0087] By implementing step 6, an energy recovery and utilization scheme is obtained. This scheme includes key energy recovery parameters such as thermoelectric conversion module configuration parameters, thermoelectric power generation parameters, output power parameters, power distribution strategy, intelligent air-cooling control parameters, and environmental temperature adaptability data, providing an energy optimization basis for the subsequent construction of the insulation system.

[0088] Step 7: Based on the energy recovery and utilization scheme, multi-layer composite insulation is carried out to obtain an optimized insulation structure scheme; Specifically, the following steps are included: Step 7.1: Design a multi-layer composite insulation system based on the box structure to obtain an insulation layer with low thermal conductivity; The enclosure needs to create a 90-degree Celsius temperature difference barrier between the external ambient temperature of 25 degrees Celsius and the internal temperature of -65 degrees Celsius. A four-layer composite structure—outer shell, vacuum insulation panel, aerogel felt, and inner wall—is employed to achieve an insulation layer with extremely low thermal conductivity. In one embodiment, the outer shell is made of aluminum alloy honeycomb panel, with a honeycomb core made of 5052 aluminum alloy. The honeycomb aperture is 10 mm, the core layer thickness is 15 mm, the outer panel thickness is 1.5 mm, and the inner panel thickness is 1 mm. The honeycomb panel has high strength and low weight characteristics, weighing only 5 kg per square meter, and has a compressive strength of 2 MPa, providing structural support for the enclosure.

[0089] A vacuum insulation panel (VIP) is attached to the inside of the outer shell. The VIP consists of a core material, a gas barrier membrane, and vacuum sealing. The core material is fumed silica with a particle size of 5 to 20 nanometers and a bulk density of 200 kg / m³, forming a nanoporous structure. The pore size is smaller than the free path of air molecules, making gas thermal conductivity negligible under vacuum conditions. The gas barrier membrane is a multi-layer composite film, consisting of three layers of polyester film, aluminum foil, and polyethylene film, heat-pressed together with a thickness of 0.1 mm, preventing the penetration of external air and water vapor. Vacuum sealing uses a heat-sealing process, evacuating to a vacuum level of less than 1 Pa. After sealing, the thermal conductivity of the VIP is as low as 0.004 W / (m·K), only one-eighth that of traditional polyurethane foam. The VIP is 20 mm thick and is constructed using multiple panels spliced ​​together to cover the inner surface of the enclosure. Vacuum insulation strips are used to fill the gaps between the panels to prevent thermal bridging.

[0090] Aerogel felt is laid inside the vacuum insulation panel. The aerogel felt is primarily composed of silica aerogel, reinforced with fibers to form a flexible felt-like material. The aerogel has a porosity of up to 95%, with pore sizes ranging from 20 to 50 nanometers, a thermal conductivity of 0.018 W / (m·K), a density of 150 kg / m³, and a thickness of 10 mm. Besides its insulation function, the aerogel felt also serves as a buffer layer, absorbing mechanical vibrations. The inner wall of the enclosure is made of 304 stainless steel plate, 0.8 mm thick. Stainless steel is characterized by low-temperature resistance, corrosion resistance, and easy cleaning. The surface is mirror-polished for high reflectivity, reducing radiative heat transfer. The total thickness of the four-layer composite structure is approximately 50 mm, plus 17 mm for the outer shell, 20 mm for the vacuum insulation panel, 10 mm for the aerogel felt, and 0.8 mm for the inner wall, totaling 47.8 mm.

[0091] Step 7.2: Calculate the heat transfer coefficient based on the insulation layer and optimize the thermal bridge treatment to obtain the heat transfer coefficient of the main body of the box. Based on a multi-layer insulation structure, a series thermal resistance model was used to calculate the overall heat transfer coefficient and optimize it for thermal bridges in the support columns, resulting in a heat transfer coefficient of 0.18 W / (m²·K) for the main body of the enclosure. Specifically, the heat transfer process includes three stages: external surface convection, multi-layer solid heat conduction, and internal surface convection. The thermal resistance is the sum of the thermal resistances of each stage. The natural convection heat transfer coefficient between the external surface and ambient air is 10 W / (m²·K), and the external surface convection thermal resistance is equal to 1 divided by the heat transfer coefficient, which equals 0.1 (m²·K) / W. The thermal resistance of solid thermal conductivity is the sum of the thermal conductivity coefficients of each layer divided by the thickness of each layer. The thermal conductivity of the aluminum alloy shell is 130 W / (m·K), but with a honeycomb structure, the equivalent thermal conductivity is corrected to 5 W / (m·K) according to the porosity. With a thickness of 17 mm, that is, 0.017 m, the thermal resistance is 0.0034 (m²·K) / W. The thermal conductivity of the vacuum insulation board is 0.004 W / (m·K), with a thickness of 0.02 m, and the thermal resistance is 5 (m²·K) / W. The thermal conductivity of the aerogel felt is 0.018 W / (m·K), with a thickness of 0.01 m, and the thermal resistance is 0.56 (m²·K) / W. The thermal conductivity of the stainless steel inner wall is 16 W / (m·K), with a thickness of 0.0008 m, and the thermal resistance is 0.00005 (m²·K) / W, which is negligible.

[0092] The natural convection heat transfer coefficient between the inner surface and the air inside the chamber is 5 W / (m²·K). The convective thermal resistance of the inner surface is equal to 1 divided by the heat transfer coefficient, which equals 0.2 (m²·K) / W. The total thermal resistance is equal to the outer surface convective thermal resistance plus the thermal conductivity of each solid layer plus the inner surface convective thermal resistance, which equals 5.8634 (m²·K) / W. The overall heat transfer coefficient is equal to 1 divided by the total thermal resistance, which equals 0.17 W / (m²·K).

[0093] However, the actual enclosure structure contains thermal bridges such as support columns. The support columns are used to connect the outer shell and the inner wall to provide structural strength. The support columns are made of glass fiber reinforced nylon material with a thermal conductivity of 0.3 W / (m·K), which is higher than that of insulation materials, forming thermal bridge channels. The diameter of the support column is 5 mm, and one support column is arranged every 100 mm inside the enclosure. The cross-sectional area of ​​the support column is pi multiplied by the square of the radius, which equals 19.6 square millimeters. The support columns account for about 0.2% of the area per unit area, and the heat transfer through the thermal bridges accounts for about 15%. Taking into account the influence of thermal bridges, the overall heat transfer coefficient of the enclosure is corrected to 0.195, which is about 0.2 W / (m²·K).

[0094] To further reduce the impact of thermal bridging, the support column was optimized by changing its cross-section from cylindrical to I-shaped, reducing the thermally conductive cross-sectional area. At the same time, an aerogel-filled cavity was set in the middle section of the support column to block heat flow. After optimization, the equivalent thermal conductivity of the support column decreased from 0.3 to 0.1 W / (m·K), the thermal bridging heat transfer ratio decreased from 15% to 5%, and the overall heat transfer coefficient of the box was finally optimized to 0.1785, approximately 0.18 W / (m²·K).

[0095] Step 7.3: Design a sealing structure based on the insulated box to obtain the final box; Based on the fact that the heat transfer coefficient of the main body of the enclosure is 0.18 W / (m²·K), a sealing design is also required for the opening parts such as the enclosure door. The final enclosure with an overall heat transfer coefficient of 0.19 W / (m²·K) is obtained through comprehensive calculation. Specifically, the top of the enclosure features an opening door measuring 600 mm x 400 mm. The door utilizes the same 4-layer composite insulation structure as the enclosure, with a thickness of 50 mm. The sealing between the door and the enclosure is crucial, employing a double-layer sealing structure. The inner sealing ring is made of silicone rubber with a Shore A hardness of 50A, a circular cross-section, and a diameter of 10 mm. Silicone rubber maintains elasticity even at low temperatures, with a temperature resistance range of -60°C to 200°C. The sealing ring is installed in a groove on the edge of the enclosure opening. When the door is closed, the sealing ring is compressed to form the first seal. The outer seal uses a PTFE sealing strip, 20 mm wide and 2 mm thick. PTFE has an extremely low coefficient of friction and excellent sealing performance. Adhered to the edge of the door, it forms a double seal with the inner silicone rubber sealing ring when the door is closed, effectively preventing air leakage.

[0096] The door locking system uses a 6-point latch, distributed at 6 positions along the door edge. When the latches are closed, a clamping force is applied to ensure uniform compression of the sealing ring, with a compression rate of approximately 30%, forming an airtight seal. Through airtightness testing, the enclosure is filled with tracer helium gas, and the helium concentration on the outer surface of the enclosure is measured. The leakage rate is less than 0.1 standard cubic centimeters per second, meeting high airtightness requirements. In addition to the door, the enclosure also features cable perforations and sensor mounting holes. Sealed joints are used at the perforations, and the joints are filled with silicone rubber sealant to ensure a tight seal.

[0097] The top door of the enclosure, due to frequent opening and difficult sealing, has a local heat transfer coefficient of approximately 0.25 W / (m²·K). The door area of ​​0.24 square meters accounts for 6% of the total surface area of ​​4 square meters. Although cable perforations and sensor mounting holes are sealed, thermal bridging still exists, resulting in a local heat transfer coefficient of approximately 0.30 W / (m²·K). The total area of ​​the perforations is approximately 0.02 square meters, accounting for 0.5% of the total surface area. Considering the combined effects of the door and perforations, the area-weighted average method is used to calculate the overall heat transfer coefficient of the enclosure: the main body area of ​​3.74 square meters has a heat transfer coefficient of 0.18 W / (m²·K), the door area of ​​0.24 square meters has a heat transfer coefficient of 0.25 W / (m²·K), and the perforation area of ​​0.02 square meters has a heat transfer coefficient of 0.30 W / (m²·K). The overall heat transfer coefficient is calculated as the sum of the heat transfer of each part divided by the total area divided by the temperature difference, resulting in an overall heat transfer coefficient of 0.19 W / (m²·K). Based on this heat transfer coefficient, the surface area of ​​the chamber is approximately 4 square meters, the temperature difference between the inside and outside is 90 degrees Celsius, and the steady-state heat transfer is equal to the heat transfer coefficient multiplied by the area multiplied by the temperature difference, which equals 68.4 watts. That is, the static heat load of the chamber is 43 watts. Adding the heat generated by the equipment inside the chamber (approximately 20 watts) and the heat generated by the tuna's respiration (approximately 10 watts), the total heat load is approximately 73 watts. The cooling capacity of the refrigeration system constructed in step 1 is 2380 watts, which is much greater than the heat load. Therefore, through the intermittent operation strategy in step 2, the refrigeration system only needs to operate for 3% of the time to maintain temperature balance. However, considering the temperature control accuracy and response speed, the actual operating time is set to 40%, leaving sufficient adjustment margin.

[0098] By implementing step 7, an optimized insulation structure scheme is obtained. This scheme includes key insulation parameters such as multi-layer composite structure parameters, heat transfer coefficient calculation results, sealing design scheme, thermal bridge optimization scheme, and heat load calculation results, thus completing the construction of the entire energy-saving ultra-low temperature cold chain transportation device for aquatic products.

[0099] An energy-saving ultra-low temperature cold chain transportation device for aquatic products is used to perform the aforementioned energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products, such as... Figure 2 As shown, it includes: The three-stage cascade cooling module is used to acquire DC power data from the aircraft cargo hold and generate an ultra-low temperature environment dataset through DC boost conversion and three-stage cascade cooling. The phase change energy storage module, based on an ultra-low temperature environment dataset, constructs an energy storage layer and controls intermittent operation to generate an optimized ultra-low temperature environment dataset; The magnetorheological vibration reduction module, based on an optimized ultra-low temperature environment dataset, performs magnetorheological vibration reduction and combines it with a collaborative control strategy to obtain a vibration reduction and protection control scheme. The multispectral quality monitoring module, based on aquatic products in the vibration reduction protection control scheme, performs defrosting-coordinated multispectral sensing and quality prediction to generate a quality monitoring dataset. The multi-parameter collaborative control module, based on the quality monitoring dataset, the optimized ultra-low temperature environment dataset, and the ultra-low temperature environment dataset, adopts a three-layer hierarchical control strategy to obtain a multi-parameter collaborative control scheme. The thermoelectric conversion energy recovery module, based on a multi-parameter collaborative control scheme, performs thermoelectric conversion to obtain an energy recovery and utilization scheme; The vacuum thermal insulation module, based on an energy recovery and utilization scheme, employs multi-layer composite insulation to achieve an optimized insulation structure.

[0100] In one embodiment of the present invention, a specific example is provided: This invention focuses on the application of high-end aquatic products in trans-Pacific air cold chain transportation. To verify the actual application effect of the device, a 30-day field test was conducted on a trans-Pacific air cold chain transportation route from Hokkaido in one country to another. During the test, five sets of energy-saving ultra-low temperature cold chain transportation devices were deployed. Each device was equipped with a complete three-stage cascade refrigeration system, phase change energy storage system, magnetorheological vibration reduction system, multispectral quality monitoring system, and layered control system. The test sample was bluefin tuna, with a single transport weight of 40 kg to 60 kg. A total of 15 transportation missions were completed, with a total transportation distance of over 15,000 kilometers.

[0101] In the initial test, a seafood import and export company purchased a batch of 50 kg bluefin tuna, which needed to be airlifted to a high-end Japanese restaurant in a certain country within 48 hours. The tuna underwent immediate pre-processing after being caught, including evisceration and cleaning, and was then loaded into the energy-saving ultra-low temperature cold chain transport device of this invention. The device parameters were set as follows: target temperature set at -65 degrees Celsius, allowable temperature fluctuation range ±0.5 degrees Celsius, myoglobin oxidation alarm threshold of 3%, and lipid peroxidation alarm threshold of 0.5 mg / kg. Before the transport process began, the device was pre-cooled on the ground, and the three-stage cascade refrigeration system was activated. After 150 minutes, the chamber temperature dropped from the ambient temperature of 25 degrees Celsius to -65 degrees Celsius. At this point, the composite phase change material completed cold storage, storing 7800 kJ of cold energy. 50 kg of tuna was placed in the container. The initial temperature of the tuna was -30 degrees Celsius. After the refrigeration system continued to run for 30 minutes, the core temperature of the tuna dropped to -65 degrees Celsius. After loading was completed, the container door was closed.

[0102] The dataset of ultra-low temperature environments obtained during the initial test transport is shown in Table 1: Table 1: Data set of ultra-low temperature environment obtained during the first test transportation;

[0103] The ultra-low temperature environment dataset shows that the three-stage cascade refrigeration system constructed in step 1 operated stably throughout the 48-hour transportation process. The chamber temperature was maintained within the range of -65 degrees Celsius ± 0.5 degrees Celsius. The temperature distribution at the eight points was uniform, with a temperature difference of less than 0.5 degrees Celsius. The frequency of the three-stage compressor was automatically adjusted according to changes in ambient air pressure. When the air pressure dropped to 75 kPa during the cruise phase, the compressor frequency was reduced through an air pressure compensation control strategy. The system energy efficiency ratio remained between 0.85 and 0.88, verifying the stability and energy efficiency of the ultra-low temperature refrigeration system.

[0104] Meanwhile, throughout the transportation process, the multispectral quality monitoring system constructed in step 4 was used to collect spectral data of the tuna surface in real time and predict quality indicators. The obtained quality monitoring dataset is shown in Table 2. Table 2: Acquired quality monitoring dataset;

[0105] The quality monitoring dataset shows that as transportation time increases, the reflectance of visible light characteristic wavelengths (R410 and R525) gradually decreases, while R630 gradually increases, indicating that myoglobin gradually oxidizes. The absorbance of near-infrared characteristic wavelengths (A1210 and A1730) gradually increases, indicating that lipid peroxidation gradually increases. The partial least squares regression prediction model established in step 4.3 calculates that the myoglobin oxidation level gradually increases from an initial 0.5% to 2.4% after 48 hours, and the lipid peroxidation value gradually increases from an initial 0.12 mg / kg to 0.40 mg / kg after 48 hours. Both quality indicators are below the alarm threshold, verifying the effectiveness of ultra-low temperature preservation. After 48 hours, the tuna myoglobin oxidation level is 2.4%, below the 3% threshold, and the lipid peroxidation value is 0.40 mg / kg, below the 0.5 mg / kg threshold, meeting sashimi-grade quality standards.

[0106] Energy consumption data shows that the total energy consumption over 48 hours was 67.2 kWh, of which the refrigeration system consumed 63 kWh, the control system consumed 3 kWh, and the thermoelectric conversion system in step 6 recovered 2.1 kWh of electricity, resulting in a net energy consumption of 65.1 kWh. Compared to traditional liquid nitrogen transportation methods, which consume approximately 100 kg of liquid nitrogen (costing 10 yuan per kg, or 1000 yuan) and electricity (costing 1 yuan per kWh, or 65 yuan), this invention's energy consumption cost is only 6.5% of the traditional method, significantly reducing transportation costs. After arriving in Shanghai, the tuna underwent sensory evaluation and physicochemical testing. It was found to be bright red in color, firm in texture, odorless, with a volatile basic nitrogen value of 8 mg / 100g and a total bacterial count of 3.2 x 10³ colony-forming units per gram, all meeting the quality standards for sashimi-grade tuna. The restaurant's purchasing party expressed satisfaction with the preservation effect.

[0107] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. An energy-saving method for ultra-low temperature cold chain transportation and preservation of aquatic products, characterized in that, Includes the following steps: Acquire DC power data from aircraft cargo holds, and generate an ultra-low temperature environment dataset through DC boost conversion and three-stage cascade cooling; Based on the ultra-low temperature environment dataset, an energy storage layer is constructed and intermittent operation is controlled to generate an optimized ultra-low temperature environment dataset; Based on the optimized ultra-low temperature environment dataset, magnetorheological vibration reduction is performed and combined with a collaborative control strategy to obtain a vibration reduction and protection control scheme. Based on the aquatic products in the vibration reduction protection control scheme, defrosting is combined with multispectral sensing and quality prediction to generate a quality monitoring dataset. Based on the quality monitoring dataset, the optimized ultra-low temperature environment dataset, and the ultra-low temperature environment dataset, a three-layer hierarchical control strategy is adopted to obtain a multi-parameter collaborative control scheme. Based on a multi-parameter collaborative control scheme, thermoelectric conversion is carried out to obtain an energy recovery and utilization scheme. Based on the energy recovery and utilization scheme, a multi-layer composite insulation scheme is adopted to obtain an optimized insulation structure.

2. The energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products according to claim 1, characterized in that, The generated ultra-low temperature environment dataset includes: The DC power supply data of the aircraft cargo hold is acquired, and the DC power supply of the aircraft cargo hold is converted into voltage using a DC boost converter module to obtain the power supply for the compressor. Design a three-stage cascade refrigeration system. The low-temperature stage refrigeration circuit uses the first refrigerant, and the evaporation temperature of the low-temperature stage evaporator is set to the first temperature threshold. The medium-temperature stage refrigeration circuit uses the second refrigerant, and the medium-temperature stage evaporator also serves as the low-temperature stage condenser. The high-temperature stage refrigeration circuit uses the third refrigerant, and the high-temperature stage evaporator also serves as the medium-temperature stage condenser. The evaporator inside the chamber is connected to the low-temperature stage refrigeration circuit, and a low-temperature axial flow fan is configured to form air circulation. Temperature sensors are arranged to monitor the temperature. The three-stage compressor starts in sequence according to a preset order, and the temperature inside the chamber drops to the second temperature threshold.

3. The energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products according to claim 1, characterized in that, The generated optimized cryogenic environment dataset includes: To formulate composite phase change materials, a low-temperature eutectic mixture is selected as the phase change substrate. A nano-nucleating agent is added to reduce the supercooling of the phase change process, and a thermal conductivity enhancer is added to improve the thermal conductivity. The phase change temperature is controlled at the third temperature threshold by adjusting the component ratio. The composite phase change material is encapsulated and installed on the inner wall of the box. A honeycomb panel is used as the encapsulation container. The composite phase change material is injected into the honeycomb cavity. Heat-conducting fins are set in the honeycomb cavity to enhance heat transfer and form a phase change energy storage layer. A phase change energy storage layer is used to control the intermittent operation of the refrigeration system. During the refrigeration stage, a three-stage compressor is run to lower the temperature of the housing and allow the phase change material to store cold energy. During the insulation stage, the compressor is turned off and the phase change material releases cold energy to maintain temperature stability, with temperature fluctuations controlled within a preset range.

4. The energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products according to claim 1, characterized in that, The obtained vibration reduction and protection control scheme includes: A magnetorheological damper is installed at the bottom of the housing, including an outer cylinder, a piston rod, a piston head, a magnetorheological fluid chamber, and an excitation coil. The apparent viscosity of the magnetorheological fluid is changed by adjusting the excitation current, and the damping coefficient is adjusted within a preset range. Vibration signals are collected using a triaxial accelerometer and the optimal damping force is calculated. A semi-active control algorithm is used to numerically integrate the acceleration signal and calculate the desired damping force according to the ceiling damping control law. The required excitation current is calculated through the vibration damper mechanical model. The root mean square value of vibration acceleration is calculated in real time. When the root mean square value of vibration exceeds the first vibration threshold, a cooling command is sent to the temperature control system to reduce the temperature of the chamber to the fourth temperature threshold. The peak value of vibration acceleration is controlled below the second vibration threshold.

5. The energy-saving method for ultra-low temperature cold chain transportation and preservation of aquatic products according to claim 1, characterized in that, The generated quality monitoring dataset includes: A multispectral sensor is installed inside the enclosure and a defrosting device is configured. The multispectral sensor includes a near-infrared spectrometer and a visible light spectrometer. The probe surface temperature is controlled by a pulse heating mode, and spectral data is collected during the defrosting window. Spectral data of aquatic product surface are collected, and a light source and spectrometer are triggered to collect reflectance spectra. The reflectance and absorbance of characteristic wavelengths are extracted to obtain spectral characteristic parameters. A partial least squares regression prediction model was adopted, with spectral characteristic parameters as independent variables and myoglobin oxidation degree and lipid peroxidation value as dependent variables, to establish a linear regression relationship and obtain real-time quality index data.

6. The energy-saving method for ultra-low temperature cold chain transportation and preservation of aquatic products according to claim 1, characterized in that, The obtained multi-parameter collaborative control scheme includes: The fuzzy control algorithm is adopted, with the input being the temperature deviation and the rate of change of the deviation inside the chamber, and the output being the compressor frequency increment, and the control cycle being within the first time threshold. A model predictive control algorithm is adopted, with inputs including temperature distribution inside the chamber, ambient air pressure and quality indicators, and outputs including compressor frequency setpoint and expansion valve opening setpoint. A system predictive model and an optimization objective function are established. A deep reinforcement learning algorithm running on a cloud server is used to establish an actuator network and an evaluator network, and the control parameters are updated through offline pre-training and online optimization.

7. The energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products according to claim 1, characterized in that, The obtained energy recovery and utilization scheme includes: Install a heat exchange copper plate, arrange thermoelectric conversion modules in series, with the hot end of the thermoelectric conversion module in contact with the copper plate and the cold end attached to a heat sink plate. When there is a temperature difference between the hot end and the cold end, the thermoelectric conversion module outputs DC voltage and power. The output voltage is stepped down and stabilized to the preset voltage by the DC converter. According to the dynamic adjustment heat dissipation strategy based on the output power of the thermoelectric conversion module, forced heat dissipation is activated when the output power is lower than the first power threshold, and forced heat dissipation is reduced or turned off when the output power recovers to above the second power threshold to maintain a stable temperature difference.

8. The energy-saving method for ultra-low temperature cold chain transportation and preservation of aquatic products according to claim 1, characterized in that, The obtained optimized insulation structure scheme includes: The design of the multi-layer composite insulation system adopts a four-layer composite structure consisting of an outer shell, a vacuum insulation panel, an aerogel felt, and an inner wall. The vacuum insulation panel consists of a core material, a gas barrier film, and vacuum sealing. The core material is fumed silica, which is evacuated to a preset vacuum level. The aerogel felt is laid on the inner side of the vacuum insulation panel. The heat transfer coefficient was calculated and the thermal bridge treatment was optimized. The overall heat transfer coefficient was calculated using a series thermal resistance model of heat transfer. The thermal bridge of the support column was optimized by changing the support column to an I-shaped cross section and setting an aerogel filling cavity. The design incorporates a sealed structure with an opening door at the top of the enclosure. A double-layer sealing structure is used between the door and the enclosure, with the inner sealing ring made of silicone rubber and the outer sealing layer made of polytetrafluoroethylene sealing tape. The door is locked using a multi-point latch, and the overall heat transfer coefficient of the enclosure is controlled below a preset threshold.

9. The energy-saving method for ultra-low temperature cold chain transportation and preservation of aquatic products according to claim 2, characterized in that, The three-stage cascade cooling system design also includes: The first refrigerant in the low-temperature stage refrigeration circuit is an azeotropic mixture with a normal boiling point below the fifth temperature threshold, which can achieve an evaporation temperature below the sixth temperature threshold. The low-temperature stage compressor, the medium-temperature stage compressor, and the high-temperature stage compressor all use DC inverter compressors. The speed is adjusted by the inverter controller. Each stage circuit is equipped with an electronic expansion valve to adjust the refrigerant flow. The valve opening adjustment range is within the preset range. A pressure sensor is installed on the outside of the enclosure to establish a pressure-based refrigeration power compensation model. When the pressure drops below the first pressure threshold, the condenser fan speed and compressor frequency are increased to compensate for the pressure effect.

10. An energy-saving ultra-low temperature cold chain transportation device for aquatic products, characterized in that, A method for implementing an energy-saving ultra-low temperature cold chain transportation and preservation method for aquatic products as described in any one of claims 1-9 includes: The three-stage cascade cooling module is used to acquire DC power data from the aircraft cargo hold and generate an ultra-low temperature environment dataset through DC boost conversion and three-stage cascade cooling. The phase change energy storage module, based on an ultra-low temperature environment dataset, constructs an energy storage layer and controls intermittent operation to generate an optimized ultra-low temperature environment dataset; The magnetorheological vibration reduction module, based on an optimized ultra-low temperature environment dataset, performs magnetorheological vibration reduction and combines it with a collaborative control strategy to obtain a vibration reduction and protection control scheme. The multispectral quality monitoring module, based on aquatic products in the vibration reduction protection control scheme, performs defrosting-coordinated multispectral sensing and quality prediction to generate a quality monitoring dataset. The multi-parameter collaborative control module, based on the quality monitoring dataset, the optimized ultra-low temperature environment dataset, and the ultra-low temperature environment dataset, adopts a three-layer hierarchical control strategy to obtain a multi-parameter collaborative control scheme. The thermoelectric conversion energy recovery module, based on a multi-parameter collaborative control scheme, performs thermoelectric conversion to obtain an energy recovery and utilization scheme; The vacuum thermal insulation module, based on an energy recovery and utilization scheme, employs multi-layer composite insulation to achieve an optimized insulation structure.

Images